Here’s where some borrowing from the field of audible acoustics might be helpful. While nearly all higher frequencies of audible sound cannot be sensed by most humans past the age of 40 or so, they are provably well qualified to differentiate between a good quality and bad quality music performance, from the right use of musical instruments to the ambient sound quality of the listening hall.
How is this possible? One answer is that even an age-deteriorated binaural hearing sense differentiates between the audio impulses arriving from different sides of the listening area, at different times, thanks to the length of bounces from the direct instrument via the walls, ceiling and rear.
The mind then assembles all these complex sounds, and quickly (this is acoustics, so it’s not exactly at supersonic speeds) generates a composite signal that is compared with memories of excellent acoustic experiences from the past, and a rating awarded, internally, although the actual frequencies can no longer be sensed.
One can conceptualise an analogue to this, where the terahertz reflections from the virus crystal are assessed in terms of much lower, and easier to sense reliably, resonant frequencies.
I’m not very familiar with radio telescopes, but it is quite obvious that these kind of problems have long been solved for star measurements, even at really high frequencies.
Since a healthcare testing gadget does not have to solve random problems of the universe, but only some specific known issues, whose parameters can be stored in a lookup table, I’m guessing that a test device can actually be made relatively inexpensively.
Not suggesting it can be done overnight either, but that it is achievable.